Académique Documents
Professionnel Documents
Culture Documents
2. Whorf and Sapir argue: We cut up natureorganize it into conceptsand ascribe significances as we do, largely because of absolutely obligatory patterns of our own language. (Whorf) The world is presented in a kaleidoscopic flux of impressions which have to be organized largely by the linguistic systems in our minds. (Whorf) Meanings are not so much discovered in experience as imposed upon it, because of the tyrannical hold that linguistic form has upon our orientation to the world. (Sapir)
B. Evidence
Four general classes of evidence are used to support the Sapir-Whorf hypothesis: 1. Languages make different distinction in their lexicons. A distinctively sculpted lexicon is the evolutionary product of a peoples struggle to survive in a specific environment. But of course the environment itself can be human-made. Latin has two words for blood: sanguis (blood inside the body), cruor (blood flowing outside the body). And if vocabulary differences are equivalent to differences in conceptual structure, then there are such differences within a given speech community: experts always have a more elaborate vocabulary for mapping a domain of experience. 2. Vocabulary differences have behavioral effects. Languages vary in the number of basic colors words they have. Many anthropologists argued that this was strong evidence for linguistic relativity: color distinctions seemed to be created by language. Rosch (1972)
2 disagreed, arguing from research among Dani people (New Guinea; two basic color terms) that speakers of all language have the same physiological ability to perceive distinctions among the 11 landmark colors. Berlin and Kay (1969) showed that these landmark colors are the basis of color systems in all world languages However, color vocabulary still provides support for linguistic relativity: the number of basic words in a persons color vocabulary influence how easy it is for that person to recognize those distinctions. The Kay-Kempton experiment (1984). This experiment involved a green-blue discrimination task. Subjects saw three color chips in the green-blue range. For example,
Phase 1: Subjects: English speakers and Tarahumara speakers (for whom there is a single word for green/blue, siyname). For 56 triads, subjects are asked: Which of the three chips is most different from the other two? English speakers picked Chip C while Tarahumaran subjects chose A or split evenly. Phase 2: Subjects: English speakers. Similar task to Phase 1, but this time subjects had a moving window through which they could only see two chips at a time. They were asked to compare the relative greenness of A and B and the relative blueness of B and C. What did they then pick for the most different chip? 3. Basic grammatical structure can differ radically from language to language. Even if we expect lexical distinctions to vary from culture to culture, we dont expect basic grammar to vary much, because it supposedly reflects fundamental distinctions like mass vs. count and entity vs. property. But it does. For example, in Native American languages, there is no clear grammatical separation between states of affairs and entities: Siberian Yupik angya -ghlla -ng boat -augmentative-acquire He wants to acquire a boat. Boat-acquiring desire Wichita (Caddoan, Oklahoma) kiya-:ki -riwa:c -e:rhira -s quotative past big buffalo They say there was a big buffalo lying there. Apparent past prone state of big buffalo -?irhawi lie -yug -tuq -desire-3SG
4. Grammatical structure has behavioral effects. For example, Navajo verb forms encode shape, flatness and flexibility of objects acted upon. Carroll and Casagrande (1958) studied what properties Navajo children use to group objects. They gave subjects a blue rope and yellow stick and asked which of the two goes with a blue stick. The
3 Navajo-speaking children chose shape (yellow stick); the English-speaking children choose color (blue rope).
4. Events that occurred outside the speakers view are reported by means of subjective marking, since there is only indirect evidence for them. This type of system is called an EVIDENTAL SYSTEM. It is common in the worlds languages. Turkish Dirseg-imi elbow 1SG.poss I hit my elbow! vur
OBJ
-du hit
-um
PST
1SG
Dirseg-imi vur elbow 1SG.POSS OBJ I must have hit my elbow! They tell me I hit my elbow.
1SG
ni -lthawakothite 1SGfork outline branch pertaining to toes (c) (d) Shawnee ni- -kwashkwi1SGrecoil ni- -kwashk1SGrecoil (e) (f) -tepe -nlocus at head hand action -a cause I push his head back. I drop it and it floats.
The boat is grounded on the beach. The boat is manned by a select group of men.
Both are statements about the boat. They are similar in this regard. But in Nootka, these two sentences are not similar. Nootka (Wakashan, Southwestern British Columbia)
5 tlih is ma moving pointwise on the beach manifest lash select tskwiq result ista ma in a canoe as a crew manifest
2. Observations There is no subject-predicate division in Nootka, contrary to Aristotles claims about the primal nature of this division. There are not even words as we know them. There is no division between nouns and verbs. Sentences are the same as words, i.e., a root plus prefixes and suffixes. (Polysynthesis) Language does not provide an unmediated picture of reality (compare Reddys paper on the conduit metaphor). [T]o restrict thinking to English...is to lose a power of thought which once lost can never be regained. It is the plainest English which contains the greatest number of unconscious assumptions about nature. (p. 244) The very thought of learning from other language patterns about ways of conceiving of the world defies a strict version of linguistic relativity.
5. The category of MATTER has two different manifestations in language: entities and reified entities (like time). (a) Reification. We can make events into countable units and abstract concepts into substances. In SAE, a cyclic phenomenon (passage of days) leads to the counting of units (10 days, etc.). In Hopi, there are no time units (or so Whorf claims). (b) Entities. Languages differ in how or whether they distinguish between MASSES and COUNTABLE OBJECTS. Despite what Whorf implies, Native American languages do distinguish between masses and countable objects (see Lakhota example below). Mass vs. count in English. *She saw cat. She saw a cat. She saw cats. She drank coffee. ? She drank waters. ? She drank a water. Mass vs. count in French: mass nouns receive the partitive article: Je voudrais encore du vin. I would like some more wine. On a mang de la choucroute. We ate some sauerkraut. Mass vs. count in Vietnamese: no distinction; all specific nouns get a CLASSIFIER. Ti an I ate I ate fish. mn dish classifier c. fish c. fish
Ti mua cn I buy animal classifier I bought a fish. Ti doc t I read paper classifier I read the newspaper.
Mass vs. count in Finnish: boundedness signaled by the noun Join kahvia drank-1SG coffee-PART I drank some coffee.
7 I was drinking coffee. Join kahvin drank-1SG coffee-ACC I drank the coffee. I a drank a cup of coffee. Mass vs. count in Czech: boundedness signaled by the verb Pil vno drink:3SG:MASC:IMPERF wine He was drinking some wine. He drank wine. Vypil vno. drink:3SG:MASC:PERF wine He drank the wine. He drank a glass of wine. Mass vs. count in Lakhota: boundedness indicated by determiner selection Mn ki blatk water def I.drank I drank the water. Mn ey blatk water some I.drank I drank some water. yelo. declarative yelo. declarative
6. The category of SPACE has two different manifestations in language: TIMESPACE METAPHORS and SPATIAL RELATIONS. (a) Hopi has no time-space metaphors according to Whorf. (b) However, it does have concepts like near and far, above and below. (c) Malotki (1983) argues against Whorfs view of Hopi: the technique of spatio-temporal metaphorization is a ubiquitous phenomenon in Hopi. It involves not only countless postpositions and adverbs of place but also a number of verbs and nouns, among them a direct equation of the noun qeni space with the notion time. (p. 15). Malotkis examples include: Nu pay tsa-y-ngahaqa-qw ya-n I ASSR small-size-INDEF-from this-way Ive been talking this way from childhood. Suu-kw taala-t One-ACC day-ACC He added one day to it. pam that a-w REF-to yua-a-ta talk-RDP-IMPRF
hoyo-k-na move-k-CAUS
7. The category of TIME has two different manifestations in language: TEMPORAL SEQUENCE and TENSE (a) Whorf observes that Hopi has no tense. (b) However, Hopi does have ways of indicating simultaneity and anteriority via subordinate clauses. (c) Whats so important about tense anyway? In Vietnamese, there are no obligatory tenses: C mt ngui dn b thuong ti tim c ph mi sang cha nht d mua c ph. Theres a lady who always comes into the caf every Sunday morning to get coffee. Ba ta rt l xu, ngu v mp nhu con heo. She is very ugly, stupid and fat, like a pig. Mt hm b ta dn v ti dang ni chuyn voi mt ngui ban lm. One day she arrived and I was chatting with a work friend. Ti hoi ban lm cua ti rang: Anh c mun cuoi mt ngui vo ngu v mp nhu b ta khng? I asked my friend, Would you want to marry someone stupid and fat like that? Bng b y nghe chng ti dang ni chuyn v b y Suddenly the woman realized that we were talking about her v b y phn-nhan cng vi ng chu cua ti. and she went to complain to my boss. [] V ng chu cua ti mun ti xin li b y And my boss wanted me to apologize to her nhung ti da khng xin li b y v ti ght b y lam. but I never did because I really hated her. In Latin, there is no one grammatical category that we could refer to as the Past tense. There are two Past forms, one for events and one for states: Marius Marius:N oppidum town:N ad Zamam to Zama:A munitum fortified:N pervenit. Id went:3g:perf:act:ind it:N erat. was:3sg:imp:act:ind
Marius went to Zama. That town was well fortified. (Sallust, Jugurtha 57.1) 8. Habitual thought in SAE and Hopi. Hopi emphasis on preparedness can be traced to the linguistic system, in which days are not distinct but successive iterations. The Hope worldview is fatalistic, as signaled by a linguistic encoding of unfolding events as manifesting.
9 The Standard Average European mentality is one in which time is quantified, allocated, measured. The TIME AS A RESOURCE metaphor is said by Whorf to stem from this mindset. Whorf doubts that a metaphor in which time is matter is primal. It stems from an industrialized culture in which measurement of time became crucial. Does Whorf believe that language influences culture? That culture influences language? There are connections but not correlations or diagnostic correspondences between cultural norms and linguistic patterns. There is a relation between a language and the rest of the culture of the society that uses it. (p. 159) Does Whorf avoid circular argumentation? That is, arguments of the following form: this grammatical form implicates this fact about the culture; this culture is one that would require this form.